{
  "id": "qwen2-5-7b-instruct",
  "identity": {
    "family": "qwen2.5",
    "display_name": "Qwen2.5 7B Instruct",
    "hub_id": "Qwen/Qwen2.5-7B-Instruct",
    "api_name": null,
    "revision": null,
    "variants": [],
    "license": "Apache-2.0",
    "commercial_use": true,
    "commercial_use_note": "Apache-2.0 weights; no use restrictions.",
    "release_date": "2024-09",
    "paper_url": "https://arxiv.org/abs/2412.15115",
    "repo_url": "https://github.com/QwenLM/Qwen2.5"
  },
  "task": {
    "category": "generation",
    "modalities_in": [
      "text"
    ],
    "modalities_out": [
      "text",
      "structured"
    ],
    "best_at": "Local structured extraction from scraped pages with no provider API calls (web_scraper llm_provider=vllm).",
    "worst_at": "Runs nowhere today (vllm is not in the engine dependencies); needs half an L4 and 6 GB per actor when it does; text only."
  },
  "io": {
    "input": {
      "text": {
        "max_tokens": "8192 (vllm_max_model_len)"
      }
    },
    "output": {
      "text": "JSON matching response_shape"
    }
  },
  "mixpeek_wiring": {
    "extractors": [
      {
        "extractor": "web_scraper@v1",
        "selected_by": "llm_provider=vllm with llm_model=Qwen/Qwen2.5-7B-Instruct (LOCAL_LLM_MODELS)"
      }
    ],
    "feature_uris": [],
    "vector_indexes": [],
    "retriever_stages": [],
    "composes_with": [
      "web_scraper crawl -> LLMExtractorProcessor on vLLM -> extracted_data"
    ]
  },
  "run": {
    "status": "not_runnable",
    "status_reason": "web_scraper's vLLM path imports vllm (shared/models/llm/vllm.py), but vllm is not in server/pyproject.toml or poetry.lock, so the import fails and the step cannot run.",
    "serve_apps": [],
    "runtime": "vllm.LLM (VLLMService, trust_remote_code false, gpu_memory_utilization 0.85)",
    "weights_format": "safetensors",
    "trust_remote_code": false,
    "size_on_disk_mb": 15231,
    "params_millions": 7616,
    "preprocessing": "Page content and title into the extraction prompt.",
    "postprocessing": "Parse JSON against response_shape.",
    "serving_shape": "batch",
    "batch_size": 12,
    "serve_batch": null,
    "scale_to_zero": {
      "min_replicas": null,
      "max_replicas": null,
      "idle_timeout_s": null,
      "memory_gb": 6,
      "num_cpus": 0.5,
      "num_gpus": 0.5,
      "cold_start_s": {
        "warm_node_s": null,
        "new_node_s": null,
        "source": null
      },
      "cpu_fallback": false
    }
  },
  "cost": {
    "unmeasured": true,
    "measured": null,
    "note": "Never run in this codebase's image; nothing measured."
  },
  "evidence": {
    "unmeasured": true,
    "items": []
  },
  "refs": [
    "hf://Qwen/Qwen2.5-7B-Instruct@a09a35458c702b33eeacc393d103063234e8bc28"
  ],
  "formats": [
    {
      "ref": "hf://Qwen/Qwen2.5-7B-Instruct@a09a35458c702b33eeacc393d103063234e8bc28",
      "format": "safetensors",
      "file": "4 files, e.g. model-00001-of-00004.safetensors",
      "loader": "transformers / open_clip from_pretrained; Mixpeek Hub import (api/models/huggingface.py)",
      "supported_today": true
    }
  ],
  "use_cases": [
    "tagging",
    "web-pages"
  ],
  "recommended": {
    "value": false,
    "for": [],
    "reason": null
  },
  "demos": [],
  "seo": {
    "title": "Qwen2.5 7B Instruct: language model",
    "description": "Qwen2.5 7B Instruct is a text language model (Apache-2.0). Not runnable on Mixpeek today.",
    "keywords": [
      "Qwen2.5 7B Instruct",
      "qwen2.5",
      "language",
      "Qwen/Qwen2.5-7B-Instruct",
      "tagging",
      "web-pages"
    ],
    "faqs": [
      {
        "q": "What is Qwen2.5 7B Instruct best at?",
        "a": "Local structured extraction from scraped pages with no provider API calls (web_scraper llm_provider=vllm)."
      },
      {
        "q": "What are the limits of Qwen2.5 7B Instruct?",
        "a": "Runs nowhere today (vllm is not in the engine dependencies); needs half an L4 and 6 GB per actor when it does; text only."
      },
      {
        "q": "Can Qwen2.5 7B Instruct be used commercially?",
        "a": "Yes. Apache-2.0 weights; no use restrictions."
      },
      {
        "q": "How does Qwen2.5 7B Instruct run on Mixpeek?",
        "a": "web_scraper's vLLM path imports vllm (shared/models/llm/vllm.py), but vllm is not in server/pyproject.toml or poetry.lock, so the import fails and the step cannot run."
      }
    ]
  },
  "last_verified": "2026-11-05",
  "sources": [
    "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct",
    "https://huggingface.co/api/models/Qwen/Qwen2.5-7B-Instruct",
    "https://arxiv.org/abs/2412.15115",
    "server/engine/extractors/web_scraper/v1/definition.py",
    "server/engine/extractors/web_scraper/v1/pipeline.py",
    "server/engine/extractors/web_scraper/v1/processors/llm_extractor.py",
    "server/shared/models/llm/vllm.py",
    "server/pyproject.toml"
  ],
  "registry": null,
  "templates": [],
  "usage": {
    "namespaces": null,
    "collections": null,
    "uncounted": true,
    "templates": 0,
    "counted_at": null,
    "source": null
  },
  "facets": {
    "status": "not_runnable",
    "category": "generation",
    "family": "qwen2.5",
    "license": "Apache-2.0",
    "commercial_use": true,
    "modalities_in": [
      "text"
    ],
    "modalities_out": [
      "text",
      "structured"
    ],
    "use_cases": [
      "tagging",
      "web-pages"
    ],
    "recommended": false,
    "release_date": "2024-09",
    "params_millions": 7616,
    "size_on_disk_mb": 15231,
    "vector_dims": null,
    "vector_distance": null,
    "weights_format": "safetensors",
    "formats": [
      "safetensors"
    ],
    "loadable_today": true,
    "trust_remote_code": false,
    "cpu_fallback": false,
    "num_gpus": 0.5,
    "memory_gb": 6,
    "min_replicas": null,
    "serving_shape": "batch",
    "served": false,
    "ref_schemes": [
      "hf"
    ],
    "has_evidence": false,
    "cost_measured": false,
    "template_count": 0,
    "usage_namespaces": null,
    "usage_collections": null,
    "usage_uncounted": true,
    "has_demo": false,
    "rank_score": 0
  },
  "rank_score": 0
}